Convolutional Neural Networks

Project: Write an Algorithm for a Dog Identification App


In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!

Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.

In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.

Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.

The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.


Why We're Here

In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).

Sample Dog Output

In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!

The Road Ahead

We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.

  • Step 0: Import Datasets
  • Step 1: Detect Humans
  • Step 2: Detect Dogs
  • Step 3: Create a CNN to Classify Dog Breeds (from Scratch)
  • Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)
  • Step 5: Write your Algorithm
  • Step 6: Test Your Algorithm

Step 0: Import Datasets

Make sure that you've downloaded the required human and dog datasets:

  • Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location /dogImages.

  • Download the human dataset. Unzip the folder and place it in the home directory, at location /lfw.

Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.

In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.

In [3]:
import numpy as np
from glob import glob

# load filenames for human and dog images
human_files = np.array(glob("lfw/*/*"))
dog_files = np.array(glob("dogImages/*/*/*"))

# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
There are 13233 total human images.
There are 8351 total dog images.

Step 1: Detect Humans

In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.

OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.

In [4]:
import cv2                
import matplotlib.pyplot as plt                        
%matplotlib inline                               

# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')

# load color (BGR) image
img = cv2.imread(human_files[0])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# find faces in image
faces = face_cascade.detectMultiScale(gray)

# print number of faces detected in the image
print('Number of faces detected:', len(faces))

# get bounding box for each detected face
for (x,y,w,h) in faces:
    # add bounding box to color image
    cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
    
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)

# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Number of faces detected: 1

Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.

In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.

Write a Human Face Detector

We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.

In [5]:
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
    img = cv2.imread(img_path)
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    faces = face_cascade.detectMultiScale(gray)
    return len(faces) > 0

(IMPLEMENTATION) Assess the Human Face Detector

Question 1: Use the code cell below to test the performance of the face_detector function.

  • What percentage of the first 100 images in human_files have a detected human face?
  • What percentage of the first 100 images in dog_files have a detected human face?

Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.

Answer: (You can print out your results and/or write your percentages in this cell)

In [6]:
from tqdm import tqdm

human_files_short = human_files[:100]
dog_files_short = dog_files[:100]

#-#-# Do NOT modify the code above this line. #-#-#

## TODO: Test the performance of the face_detector algorithm 
## on the images in human_files_short and dog_files_short.
#print('first file name is', human_files[0]) 
##the above will output lfw\Aaron_Eckhart\Aaron_Eckhart_0001.jpg in my running
n=100 #100 images
##the following process for the human_files_short
detect_human=[face_detector(img_file) for img_file in tqdm(human_files_short)]
detect_human_count=sum(detect_human)
print('Out of the first ',n, ' images in human_files ', detect_human_count, ' have face detected')
print('The (correctly) Detected Percentage in human_files_short: ',detect_human_count/n*100, '%')

##the following process for the dog_files_short
detect_dog=[face_detector(img_file) for img_file in tqdm(dog_files_short)]
detect_dog_count=sum(detect_dog)
print('Out of the first ',n, ' images in dog_files ', detect_dog_count, ' have face detected')
print('The (wrongly) Detected Percentage in dog_files_short: ',detect_dog_count/n*100, '%')
100%|██████████| 100/100 [00:02<00:00, 38.19it/s]
  0%|          | 0/100 [00:00<?, ?it/s]
Out of the first  100  images in human_files  99  have face detected
The (correctly) Detected Percentage in human_files_short:  99.0 %
100%|██████████| 100/100 [00:11<00:00,  8.58it/s]
Out of the first  100  images in dog_files  18  have face detected
The (wrongly) Detected Percentage in dog_files_short:  18.0 %

We suggest the face detector from OpenCV as a potential way to detect human images in your algorithm, but you are free to explore other approaches, especially approaches that make use of deep learning :). Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.

In [7]:
### (Optional) 
### TODO: Test performance of another face detection algorithm.
### Feel free to use as many code cells as needed.
### use an alternative Haar feature-based cascade classifier download from opencv github 
alter_xml_file='haarcascade_frontalface_alt2.xml'
face_cascade2 = cv2.CascadeClassifier('haarcascades/' + alter_xml_file)
def face_detector2(img_path):
    img = cv2.imread(img_path)
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    faces = face_cascade2.detectMultiScale(gray)
    return len(faces) > 0

print('***The following uses', alter_xml_file, 'classifier****')

detect_human=[face_detector2(img_file) for img_file in tqdm(human_files_short)]
detect_human_count=sum(detect_human)
print('Out of the first ',n, ' images in human_files ', detect_human_count, ' have face detected')
print('The (correctly) Detected Percentage in human_files_short: ',detect_human_count/n*100, '%')

##the following process for the dog_files_short
detect_dog=[face_detector2(img_file) for img_file in tqdm(dog_files_short)]
detect_dog_count=sum(detect_dog)
print('Out of the first ',n, ' images in dog_files ', detect_dog_count, ' have face detected')
print('The (wrongly) Detected Percentage in dog_files_short: ',detect_dog_count/n*100, '%')
  4%|▍         | 4/100 [00:00<00:02, 36.94it/s]
***The following uses haarcascade_frontalface_alt2.xml classifier****
100%|██████████| 100/100 [00:02<00:00, 36.61it/s]
  0%|          | 0/100 [00:00<?, ?it/s]
Out of the first  100  images in human_files  99  have face detected
The (correctly) Detected Percentage in human_files_short:  99.0 %
100%|██████████| 100/100 [00:13<00:00,  7.20it/s]
Out of the first  100  images in dog_files  21  have face detected
The (wrongly) Detected Percentage in dog_files_short:  21.0 %


Step 2: Detect Dogs

In this section, we use a pre-trained model to detect dogs in images.

Obtain Pre-trained VGG-16 Model

The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.

In [8]:
import torch
import torchvision.models as models

# define VGG16 model
VGG16 = models.vgg16(pretrained=True, progress=False)

# check if CUDA is available
use_cuda = torch.cuda.is_available()

# move model to GPU if CUDA is available
if use_cuda:
    VGG16 = VGG16.cuda()

Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.

(IMPLEMENTATION) Making Predictions with a Pre-trained Model

In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.

Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.

In [9]:
from PIL import Image
import torchvision.transforms as transforms

# Set PIL to be tolerant of image files that are truncated.
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True

def VGG16_predict(img_path):
    '''
    Use pre-trained VGG-16 model to obtain index corresponding to 
    predicted ImageNet class for image at specified path
    
    Args:
        img_path: path to an image
        
    Returns:
        Index corresponding to VGG-16 model's prediction
    '''
    
    ## TODO: Complete the function.
    ## Load and pre-process an image from the given img_path
    ## Return the *index* of the predicted class for that image
    #read the image and convert it to RGB
    img = Image.open(img_path).convert('RGB')
    '''
    resize to (224, 224) because vgg16 requires that, according to section 2.1
    of the paper 'Very Deep Convolutional Networks for Large-Scale Image 
    Recognition' by Simonyan and Zisserman.
    '''
    img_trans = transforms.Compose([
                        transforms.Resize(size=(224, 224)),
                        transforms.ToTensor()])
    img = img_trans(img)[:3,:,:].unsqueeze(0)    
    if use_cuda: #works only when cuda is available. Currently it won't work without Nvidia Card
        img = img.cuda()
    ret = VGG16(img)
    return torch.max(ret,1)[1].item() # predicted class index    
In [10]:
# try the first dog file, and test run for the VGG16_predict
temp_test=VGG16_predict(dog_files[1])
print('predicted index of file', dog_files[1], '(using vgg16) is', temp_test)
predicted index of file dogImages/valid/122.Pointer/Pointer_07826.jpg (using vgg16) is 242

(IMPLEMENTATION) Write a Dog Detector

While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).

Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).

In [11]:
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
    ## TODO: Complete the function.
    dog_index = VGG16_predict(img_path)
    return dog_index >= 151 and dog_index <= 268 #True if between 151-268, inclusive
    return None # true/false
In [12]:
print('Testing file', dog_files[1],'. A dog?', dog_detector(dog_files[1]))
Testing file dogImages/valid/122.Pointer/Pointer_07826.jpg . A dog? True

(IMPLEMENTATION) Assess the Dog Detector

Question 2: Use the code cell below to test the performance of your dog_detector function.

  • What percentage of the images in human_files_short have a detected dog?
  • What percentage of the images in dog_files_short have a detected dog?

Answer: 0% human_files_short have detected dog and 100% dog_file_short detected dog.

In [13]:
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.
def dog_detector_batch_test(files):
    detect_dog=[int(dog_detector(img_file)) for img_file in tqdm(files)]
    detect_dog_count=sum(detect_dog)
    total_count=len(files)
    return detect_dog_count, total_count

human_test_result=dog_detector_batch_test(human_files_short)
print('dog detection in human_files_short:', human_test_result[0]/human_test_result[1]*100,'%')

dog_test_result=dog_detector_batch_test(dog_files_short)
print('dog detection in dog_files_short:', dog_test_result[0]/dog_test_result[1]*100,'%')
100%|██████████| 100/100 [00:28<00:00,  3.74it/s]
  0%|          | 0/100 [00:00<?, ?it/s]
dog detection in human_files_short: 0.0 %
100%|██████████| 100/100 [00:28<00:00,  3.15it/s]
dog detection in dog_files_short: 92.0 %

We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.

In [ ]:
### (Optional) 
### TODO: Report the performance of another pre-trained network.
### Feel free to use as many code cells as needed.

Step 3: Create a CNN to Classify Dog Breeds (from Scratch)

Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.

We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.

Brittany Welsh Springer Spaniel

It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).

Curly-Coated Retriever American Water Spaniel

Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.

Yellow Labrador Chocolate Labrador Black Labrador

We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.

Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!

(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset

Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!

In [14]:
import os
from torchvision import datasets
import torchvision.transforms as transforms
import torch
import numpy as np
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True

### TODO: Write data loaders for training, validation, and test sets
## Specify appropriate transforms, and batch_sizes

batch_size = 20
num_workers = 0

data_dir = 'dogImages/'
train_dir = os.path.join(data_dir, 'train/')
valid_dir = os.path.join(data_dir, 'valid/')
test_dir = os.path.join(data_dir, 'test/')
In [15]:
standard_normalization = transforms.Normalize(mean=[0.485, 0.456, 0.406],
                                              std=[0.229, 0.224, 0.225])
In [16]:
data_transforms = {'train': transforms.Compose([transforms.RandomResizedCrop(224),
                                     transforms.RandomHorizontalFlip(),
                                     transforms.ToTensor(),
                                     standard_normalization]),
                   'val': transforms.Compose([transforms.Resize(256),
                                     transforms.CenterCrop(224),
                                     transforms.ToTensor(),
                                     standard_normalization]),
                   'test': transforms.Compose([transforms.Resize(size=(224,224)),
                                     transforms.ToTensor(), 
                                     standard_normalization])
                  }
In [17]:
train_data = datasets.ImageFolder(train_dir, transform=data_transforms['train'])
valid_data = datasets.ImageFolder(valid_dir, transform=data_transforms['val'])
test_data = datasets.ImageFolder(test_dir, transform=data_transforms['test'])
In [18]:
train_loader = torch.utils.data.DataLoader(train_data,
                                           batch_size=batch_size, 
                                           num_workers=num_workers,
                                           shuffle=True)
valid_loader = torch.utils.data.DataLoader(valid_data,
                                           batch_size=batch_size, 
                                           num_workers=num_workers,
                                           shuffle=False)
test_loader = torch.utils.data.DataLoader(test_data,
                                           batch_size=batch_size, 
                                           num_workers=num_workers,
                                           shuffle=False)
loaders_scratch = {
    'train': train_loader,
    'valid': valid_loader,
    'test': test_loader
}

Question 3: Describe your chosen procedure for preprocessing the data.

  • How does your code resize the images (by cropping, stretching, etc)? What size did you pick for the input tensor, and why?
  • Did you decide to augment the dataset? If so, how (through translations, flips, rotations, etc)? If not, why not?

Answer: I've applied RandomResizedCrop & RandomHorizontalFlip to just train_data. This will do both image augmentations and resizing jobs. Image augmentation will give randomness to the dataset so, it prevents overfitting and I can expect better performance of model when it's predicting toward test_data. On the other hand, I've done Resize of (256) and then, center crop to make 224 X 224. Since valid_data will be used for validation check, I will not do image augmentations. For the test_data, I've applied only image resizing.

(IMPLEMENTATION) Model Architecture

Create a CNN to classify dog breed. Use the template in the code cell below.

In [19]:
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True

num_classes = 133 # total classes of dog breeds
In [20]:
import torch.nn as nn
import torch.nn.functional as F
import numpy as np

# define the CNN architecture
class Net(nn.Module):
    ### TODO: choose an architecture, and complete the class
    def __init__(self):
        super(Net, self).__init__()
        ## Define layers of a CNN
        self.conv1 = nn.Conv2d(3, 32, 3, stride=2, padding=1)
        self.conv2 = nn.Conv2d(32, 64, 3, stride=2, padding=1)
        self.conv3 = nn.Conv2d(64, 128, 3, padding=1)

        # pool
        self.pool = nn.MaxPool2d(2, 2)
        
        # fully-connected
        self.fc1 = nn.Linear(7*7*128, 500)
        self.fc2 = nn.Linear(500, num_classes) 
        
        # drop-out
        self.dropout = nn.Dropout(0.3)
    
    def forward(self, x):
        ## Define forward behavior
        x = F.relu(self.conv1(x))
        x = self.pool(x)
        x = F.relu(self.conv2(x))
        x = self.pool(x)
        x = F.relu(self.conv3(x))
        x = self.pool(x)
        
        # flatten
        x = x.view(-1, 7*7*128)
        
        x = self.dropout(x)
        x = F.relu(self.fc1(x))
        
        x = self.dropout(x)
        x = self.fc2(x)
        return x

#-#-# You so NOT have to modify the code below this line. #-#-#

# instantiate the CNN
model_scratch = Net()
print(model_scratch)

# move tensors to GPU if CUDA is available
if use_cuda:
    model_scratch.cuda()
Net(
  (conv1): Conv2d(3, 32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))
  (conv2): Conv2d(32, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1))
  (conv3): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
  (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  (fc1): Linear(in_features=6272, out_features=500, bias=True)
  (fc2): Linear(in_features=500, out_features=133, bias=True)
  (dropout): Dropout(p=0.3)
)

Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step.

Answer: (conv1): Conv2d(3, 32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) activation: relu (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) activation: relu (conv2): Conv2d(32, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1)) activation: relu (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) (conv3): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1)) (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False) (dropout): Dropout(p=0.3) (fc1): Linear(in_features=6272, out_features=500, bias=True) (dropout): Dropout(p=0.3) (fc2): Linear(in_features=500, out_features=133, bias=True) explanations First 2 conv layers I've applied kernel_size of 3 with stride 2, this will lead to downsize of input image by 2. after 2 conv layers, maxpooling with stride 2 is placed and this will lead to downsize of input image by 2. The 3rd conv layers is consist of kernel_size of 3 with stride 1, and this will not reduce input image. after final maxpooling with stride 2, the total output image size is downsized by factor of 32 and the depth will be 128. I've applied dropout of 0.3 in order to prevent overfitting. Fully-connected layer is placed and then, 2nd fully-connected layer is intended to produce final output_size which predicts classes of breeds.

(IMPLEMENTATION) Specify Loss Function and Optimizer

Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_scratch, and the optimizer as optimizer_scratch below.

In [21]:
import torch.optim as optim

### TODO: select loss function
criterion_scratch = nn.CrossEntropyLoss()

### TODO: select optimizer
optimizer_scratch = optim.SGD(model_scratch.parameters(), lr = 0.05)

(IMPLEMENTATION) Train and Validate the Model

Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_scratch.pt'.

In [24]:
def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path, last_validation_loss=None):
    """returns trained model"""
    # initialize tracker for minimum validation loss
    if last_validation_loss is not None:
        valid_loss_min = last_validation_loss
    else:
        valid_loss_min = np.Inf
    
    for epoch in range(1, n_epochs+1):
        # initialize variables to monitor training and validation loss
        train_loss = 0.0
        valid_loss = 0.0
        
        ###################
        # train the model #
        ###################
        model.train()
        for batch_idx, (data, target) in enumerate(loaders['train']):
            # move to GPU
            if use_cuda:
                data, target = data.cuda(), target.cuda()
            ## find the loss and update the model parameters accordingly
            ## record the average training loss, using something like
            ## train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))

            # initialize weights to zero
            optimizer.zero_grad()
            
            output = model(data)
            
            # calculate loss
            loss = criterion(output, target)
            
            # back prop
            loss.backward()
            
            # grad
            optimizer.step()
            
            train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
            
            if batch_idx % 100 == 0:
                print('Epoch %d, Batch %d loss: %.6f' %
                  (epoch, batch_idx + 1, train_loss))
            
        ######################    
        # validate the model #
        ######################
        model.eval()
        for batch_idx, (data, target) in enumerate(loaders['valid']):
            # move to GPU
            if use_cuda:
                data, target = data.cuda(), target.cuda()
            ## update the average validation loss
            output = model(data)
            loss = criterion(output, target)
            valid_loss = valid_loss + ((1 / (batch_idx + 1)) * (loss.data - valid_loss))

            
        # print training/validation statistics 
        print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
            epoch, 
            train_loss,
            valid_loss
            ))
        
        ## TODO: save the model if validation loss has decreased
        if valid_loss < valid_loss_min:
            torch.save(model.state_dict(), save_path)
            print('Validation loss decreased ({:.6f} --> {:.6f}).  Saving model ...'.format(
            valid_loss_min,
            valid_loss))
            valid_loss_min = valid_loss
            
    # return trained model
    return model
In [25]:
# train the model
model_scratch = train(10, loaders_scratch, model_scratch, optimizer_scratch, 
                      criterion_scratch, use_cuda, 'saved_models/model_scratch.pt')
Epoch 1, Batch 1 loss: 4.900301
Epoch 1, Batch 101 loss: 4.884957
Epoch 1, Batch 201 loss: 4.881713
Epoch 1, Batch 301 loss: 4.875326
Epoch: 1 	Training Loss: 4.874252 	Validation Loss: 4.821206
Validation loss decreased (inf --> 4.821206).  Saving model ...
Epoch 2, Batch 1 loss: 4.903143
Epoch 2, Batch 101 loss: 4.814694
Epoch 2, Batch 201 loss: 4.796427
Epoch 2, Batch 301 loss: 4.791905
Epoch: 2 	Training Loss: 4.788963 	Validation Loss: 4.667059
Validation loss decreased (4.821206 --> 4.667059).  Saving model ...
Epoch 3, Batch 1 loss: 4.660416
Epoch 3, Batch 101 loss: 4.678959
Epoch 3, Batch 201 loss: 4.672832
Epoch 3, Batch 301 loss: 4.660891
Epoch: 3 	Training Loss: 4.657626 	Validation Loss: 4.466977
Validation loss decreased (4.667059 --> 4.466977).  Saving model ...
Epoch 4, Batch 1 loss: 4.597161
Epoch 4, Batch 101 loss: 4.581704
Epoch 4, Batch 201 loss: 4.579004
Epoch 4, Batch 301 loss: 4.576756
Epoch: 4 	Training Loss: 4.578728 	Validation Loss: 4.414859
Validation loss decreased (4.466977 --> 4.414859).  Saving model ...
Epoch 5, Batch 1 loss: 4.467006
Epoch 5, Batch 101 loss: 4.533636
Epoch 5, Batch 201 loss: 4.526348
Epoch 5, Batch 301 loss: 4.525908
Epoch: 5 	Training Loss: 4.522493 	Validation Loss: 4.343369
Validation loss decreased (4.414859 --> 4.343369).  Saving model ...
Epoch 6, Batch 1 loss: 4.205079
Epoch 6, Batch 101 loss: 4.445117
Epoch 6, Batch 201 loss: 4.457407
Epoch 6, Batch 301 loss: 4.455827
Epoch: 6 	Training Loss: 4.448840 	Validation Loss: 4.234025
Validation loss decreased (4.343369 --> 4.234025).  Saving model ...
Epoch 7, Batch 1 loss: 4.562132
Epoch 7, Batch 101 loss: 4.407517
Epoch 7, Batch 201 loss: 4.411787
Epoch 7, Batch 301 loss: 4.387107
Epoch: 7 	Training Loss: 4.389118 	Validation Loss: 4.189588
Validation loss decreased (4.234025 --> 4.189588).  Saving model ...
Epoch 8, Batch 1 loss: 4.231719
Epoch 8, Batch 101 loss: 4.365149
Epoch 8, Batch 201 loss: 4.361694
Epoch 8, Batch 301 loss: 4.351868
Epoch: 8 	Training Loss: 4.347396 	Validation Loss: 4.081192
Validation loss decreased (4.189588 --> 4.081192).  Saving model ...
Epoch 9, Batch 1 loss: 4.287886
Epoch 9, Batch 101 loss: 4.292417
Epoch 9, Batch 201 loss: 4.270149
Epoch 9, Batch 301 loss: 4.266828
Epoch: 9 	Training Loss: 4.265602 	Validation Loss: 4.041091
Validation loss decreased (4.081192 --> 4.041091).  Saving model ...
Epoch 10, Batch 1 loss: 4.437462
Epoch 10, Batch 101 loss: 4.239433
Epoch 10, Batch 201 loss: 4.227575
Epoch 10, Batch 301 loss: 4.227765
Epoch: 10 	Training Loss: 4.229860 	Validation Loss: 4.026369
Validation loss decreased (4.041091 --> 4.026369).  Saving model ...

(IMPLEMENTATION) Test the Model

Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 10%.

In [26]:
def test(loaders, model, criterion, use_cuda):

    # monitor test loss and accuracy
    test_loss = 0.
    correct = 0.
    total = 0.

    model.eval()
    for batch_idx, (data, target) in enumerate(loaders['test']):
        # move to GPU
        if use_cuda:
            data, target = data.cuda(), target.cuda()
        # forward pass: compute predicted outputs by passing inputs to the model
        output = model(data)
        # calculate the loss
        loss = criterion(output, target)
        # update average test loss 
        test_loss = test_loss + ((1 / (batch_idx + 1)) * (loss.data - test_loss))
        # convert output probabilities to predicted class
        pred = output.data.max(1, keepdim=True)[1]
        # compare predictions to true label
        correct += np.sum(np.squeeze(pred.eq(target.data.view_as(pred))).cpu().numpy())
        total += data.size(0)
            
    print('Test Loss: {:.6f}\n'.format(test_loss))

    print('\nTest Accuracy: %2d%% (%2d/%2d)' % (
        100. * correct / total, correct, total))

# call test function    
test(loaders_scratch, model_scratch, criterion_scratch, use_cuda)
Test Loss: 4.120332


Test Accuracy:  8% (68/836)

Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)

You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.

(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset

Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively).

If you like, you are welcome to use the same data loaders from the previous step, when you created a CNN from scratch.

In [27]:
## TODO: Specify data loaders
loaders_transfer = loaders_scratch.copy()

(IMPLEMENTATION) Model Architecture

Use transfer learning to create a CNN to classify dog breed. Use the code cell below, and save your initialized model as the variable model_transfer.

In [31]:
import torchvision.models as models
import torch.nn as nn

## TODO: Specify model architecture 
model_transfer = models.resnet50(pretrained=True)
In [32]:
for param in model_transfer.parameters():
    param.requires_grad = False
In [33]:
model_transfer.fc = nn.Linear(2048, 133, bias=True)
In [34]:
fc_parameters = model_transfer.fc.parameters()
In [ ]:
for param in fc_parameters:
    param.requires_grad = True
In [35]:
model_transfer
Out[35]:
ResNet(
  (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (relu): ReLU(inplace)
  (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
  (layer1): Sequential(
    (0): Bottleneck(
      (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
      (downsample): Sequential(
        (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): Bottleneck(
      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (2): Bottleneck(
      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
  )
  (layer2): Sequential(
    (0): Bottleneck(
      (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
      (downsample): Sequential(
        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (2): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (3): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
  )
  (layer3): Sequential(
    (0): Bottleneck(
      (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
      (downsample): Sequential(
        (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (2): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (3): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (4): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (5): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
  )
  (layer4): Sequential(
    (0): Bottleneck(
      (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
      (downsample): Sequential(
        (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): Bottleneck(
      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (2): Bottleneck(
      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
  )
  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))
  (fc): Linear(in_features=2048, out_features=133, bias=True)
)
In [36]:
if use_cuda:
    model_transfer = model_transfer.cuda()

Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem

Answer: ResNet is used as a transfer model because it is an outstanding tool on Image Classification. I looked into the structure and functions of ResNet. The core idea of ResNet is introducing a so-called “identity shortcut connection” that skips one or more layers and may prevent overfitting during training.

(IMPLEMENTATION) Specify Loss Function and Optimizer

Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_transfer, and the optimizer as optimizer_transfer below.

In [37]:
criterion_transfer = nn.CrossEntropyLoss()
optimizer_transfer = optim.SGD(model_transfer.fc.parameters(), lr=0.001)

(IMPLEMENTATION) Train and Validate the Model

Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_transfer.pt'.

In [38]:
# train the model
# train(n_epochs, loaders_transfer, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'model_transfer.pt')

def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
    """returns trained model"""
    # initialize tracker for minimum validation loss
    valid_loss_min = np.Inf
    
    for epoch in range(1, n_epochs+1):
        # initialize variables to monitor training and validation loss
        train_loss = 0.0
        valid_loss = 0.0
        
        ###################
        # train the model #
        ###################
        model.train()
        for batch_idx, (data, target) in enumerate(loaders['train']):
            # move to GPU
            if use_cuda:
                data, target = data.cuda(), target.cuda()

            # initialize weights to zero
            optimizer.zero_grad()
            
            output = model(data)
            
            # calculate loss
            loss = criterion(output, target)
            
            # back prop
            loss.backward()
            
            # grad
            optimizer.step()
            
            train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
            
            if batch_idx % 100 == 0:
                print('Epoch %d, Batch %d loss: %.6f' %
                  (epoch, batch_idx + 1, train_loss))
        
        ######################    
        # validate the model #
        ######################
        model.eval()
        for batch_idx, (data, target) in enumerate(loaders['valid']):
            # move to GPU
            if use_cuda:
                data, target = data.cuda(), target.cuda()
            ## update the average validation loss
            output = model(data)
            loss = criterion(output, target)
            valid_loss = valid_loss + ((1 / (batch_idx + 1)) * (loss.data - valid_loss))

            
        # print training/validation statistics 
        print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
            epoch, 
            train_loss,
            valid_loss
            ))
        
        ## TODO: save the model if validation loss has decreased
        if valid_loss < valid_loss_min:
            torch.save(model.state_dict(), save_path)
            print('Validation loss decreased ({:.6f} --> {:.6f}).  Saving model ...'.format(
            valid_loss_min,
            valid_loss))
            valid_loss_min = valid_loss
            
    # return trained model
    return model
In [39]:
train(20, loaders_transfer, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'saved_models/model_transfer.pt')
Epoch 1, Batch 1 loss: 4.816292
Epoch 1, Batch 101 loss: 4.925079
Epoch 1, Batch 201 loss: 4.886065
Epoch 1, Batch 301 loss: 4.852875
Epoch: 1 	Training Loss: 4.840649 	Validation Loss: 4.656911
Validation loss decreased (inf --> 4.656911).  Saving model ...
Epoch 2, Batch 1 loss: 4.622720
Epoch 2, Batch 101 loss: 4.679409
Epoch 2, Batch 201 loss: 4.651557
Epoch 2, Batch 301 loss: 4.625751
Epoch: 2 	Training Loss: 4.617173 	Validation Loss: 4.402648
Validation loss decreased (4.656911 --> 4.402648).  Saving model ...
Epoch 3, Batch 1 loss: 4.476030
Epoch 3, Batch 101 loss: 4.484774
Epoch 3, Batch 201 loss: 4.462270
Epoch 3, Batch 301 loss: 4.434701
Epoch: 3 	Training Loss: 4.422999 	Validation Loss: 4.161709
Validation loss decreased (4.402648 --> 4.161709).  Saving model ...
Epoch 4, Batch 1 loss: 4.466436
Epoch 4, Batch 101 loss: 4.290622
Epoch 4, Batch 201 loss: 4.269002
Epoch 4, Batch 301 loss: 4.242071
Epoch: 4 	Training Loss: 4.234505 	Validation Loss: 3.945987
Validation loss decreased (4.161709 --> 3.945987).  Saving model ...
Epoch 5, Batch 1 loss: 4.074571
Epoch 5, Batch 101 loss: 4.102571
Epoch 5, Batch 201 loss: 4.097570
Epoch 5, Batch 301 loss: 4.069326
Epoch: 5 	Training Loss: 4.063991 	Validation Loss: 3.730982
Validation loss decreased (3.945987 --> 3.730982).  Saving model ...
Epoch 6, Batch 1 loss: 3.991749
Epoch 6, Batch 101 loss: 3.963288
Epoch 6, Batch 201 loss: 3.933837
Epoch 6, Batch 301 loss: 3.904181
Epoch: 6 	Training Loss: 3.897418 	Validation Loss: 3.517046
Validation loss decreased (3.730982 --> 3.517046).  Saving model ...
Epoch 7, Batch 1 loss: 4.120534
Epoch 7, Batch 101 loss: 3.777073
Epoch 7, Batch 201 loss: 3.764479
Epoch 7, Batch 301 loss: 3.741257
Epoch: 7 	Training Loss: 3.731467 	Validation Loss: 3.330152
Validation loss decreased (3.517046 --> 3.330152).  Saving model ...
Epoch 8, Batch 1 loss: 3.789134
Epoch 8, Batch 101 loss: 3.626654
Epoch 8, Batch 201 loss: 3.619787
Epoch 8, Batch 301 loss: 3.599861
Epoch: 8 	Training Loss: 3.594640 	Validation Loss: 3.134459
Validation loss decreased (3.330152 --> 3.134459).  Saving model ...
Epoch 9, Batch 1 loss: 3.507753
Epoch 9, Batch 101 loss: 3.506901
Epoch 9, Batch 201 loss: 3.484256
Epoch 9, Batch 301 loss: 3.448965
Epoch: 9 	Training Loss: 3.444638 	Validation Loss: 2.976092
Validation loss decreased (3.134459 --> 2.976092).  Saving model ...
Epoch 10, Batch 1 loss: 3.420999
Epoch 10, Batch 101 loss: 3.329292
Epoch 10, Batch 201 loss: 3.312332
Epoch 10, Batch 301 loss: 3.296829
Epoch: 10 	Training Loss: 3.297010 	Validation Loss: 2.825586
Validation loss decreased (2.976092 --> 2.825586).  Saving model ...
Epoch 11, Batch 1 loss: 3.061508
Epoch 11, Batch 101 loss: 3.222434
Epoch 11, Batch 201 loss: 3.205875
Epoch 11, Batch 301 loss: 3.178463
Epoch: 11 	Training Loss: 3.173427 	Validation Loss: 2.644595
Validation loss decreased (2.825586 --> 2.644595).  Saving model ...
Epoch 12, Batch 1 loss: 2.996167
Epoch 12, Batch 101 loss: 3.089199
Epoch 12, Batch 201 loss: 3.083975
Epoch 12, Batch 301 loss: 3.064475
Epoch: 12 	Training Loss: 3.062380 	Validation Loss: 2.546844
Validation loss decreased (2.644595 --> 2.546844).  Saving model ...
Epoch 13, Batch 1 loss: 2.877346
Epoch 13, Batch 101 loss: 3.001435
Epoch 13, Batch 201 loss: 2.978536
Epoch 13, Batch 301 loss: 2.962938
Epoch: 13 	Training Loss: 2.954036 	Validation Loss: 2.401006
Validation loss decreased (2.546844 --> 2.401006).  Saving model ...
Epoch 14, Batch 1 loss: 3.251268
Epoch 14, Batch 101 loss: 2.867735
Epoch 14, Batch 201 loss: 2.854025
Epoch 14, Batch 301 loss: 2.845936
Epoch: 14 	Training Loss: 2.849858 	Validation Loss: 2.336155
Validation loss decreased (2.401006 --> 2.336155).  Saving model ...
Epoch 15, Batch 1 loss: 2.620637
Epoch 15, Batch 101 loss: 2.797500
Epoch 15, Batch 201 loss: 2.781451
Epoch 15, Batch 301 loss: 2.773891
Epoch: 15 	Training Loss: 2.769301 	Validation Loss: 2.175406
Validation loss decreased (2.336155 --> 2.175406).  Saving model ...
Epoch 16, Batch 1 loss: 2.615082
Epoch 16, Batch 101 loss: 2.661330
Epoch 16, Batch 201 loss: 2.655060
Epoch 16, Batch 301 loss: 2.654689
Epoch: 16 	Training Loss: 2.651234 	Validation Loss: 2.061422
Validation loss decreased (2.175406 --> 2.061422).  Saving model ...
Epoch 17, Batch 1 loss: 2.683201
Epoch 17, Batch 101 loss: 2.622040
Epoch 17, Batch 201 loss: 2.601982
Epoch 17, Batch 301 loss: 2.590708
Epoch: 17 	Training Loss: 2.590794 	Validation Loss: 2.028361
Validation loss decreased (2.061422 --> 2.028361).  Saving model ...
Epoch 18, Batch 1 loss: 3.232593
Epoch 18, Batch 101 loss: 2.570632
Epoch 18, Batch 201 loss: 2.531232
Epoch 18, Batch 301 loss: 2.511075
Epoch: 18 	Training Loss: 2.503981 	Validation Loss: 1.923702
Validation loss decreased (2.028361 --> 1.923702).  Saving model ...
Epoch 19, Batch 1 loss: 2.775020
Epoch 19, Batch 101 loss: 2.455524
Epoch 19, Batch 201 loss: 2.425761
Epoch 19, Batch 301 loss: 2.417483
Epoch: 19 	Training Loss: 2.417068 	Validation Loss: 1.860032
Validation loss decreased (1.923702 --> 1.860032).  Saving model ...
Epoch 20, Batch 1 loss: 2.683101
Epoch 20, Batch 101 loss: 2.376293
Epoch 20, Batch 201 loss: 2.384962
Epoch 20, Batch 301 loss: 2.383114
Epoch: 20 	Training Loss: 2.383381 	Validation Loss: 1.756267
Validation loss decreased (1.860032 --> 1.756267).  Saving model ...
Out[39]:
ResNet(
  (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (relu): ReLU(inplace)
  (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
  (layer1): Sequential(
    (0): Bottleneck(
      (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
      (downsample): Sequential(
        (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): Bottleneck(
      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (2): Bottleneck(
      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
  )
  (layer2): Sequential(
    (0): Bottleneck(
      (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
      (downsample): Sequential(
        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (2): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (3): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
  )
  (layer3): Sequential(
    (0): Bottleneck(
      (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
      (downsample): Sequential(
        (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (2): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (3): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (4): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (5): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
  )
  (layer4): Sequential(
    (0): Bottleneck(
      (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
      (downsample): Sequential(
        (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): Bottleneck(
      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
    (2): Bottleneck(
      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace)
    )
  )
  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))
  (fc): Linear(in_features=2048, out_features=133, bias=True)
)

(IMPLEMENTATION) Test the Model

Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 60%.

In [41]:
test(loaders_transfer, model_transfer, criterion_transfer, use_cuda)
Test Loss: 1.824126


Test Accuracy: 72% (610/836)

(IMPLEMENTATION) Predict Dog Breed with the Model

Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan hound, etc) that is predicted by your model.

In [42]:
### TODO: Write a function that takes a path to an image as input
### and returns the dog breed that is predicted by the model.

# list of class names by index, i.e. a name can be accessed like class_names[0]
class_names = [item[4:].replace("_", " ") for item in loaders_transfer['train'].dataset.classes]
loaders_transfer['train'].dataset.classes[:10]
Out[42]:
['001.Affenpinscher',
 '002.Afghan_hound',
 '003.Airedale_terrier',
 '004.Akita',
 '005.Alaskan_malamute',
 '006.American_eskimo_dog',
 '007.American_foxhound',
 '008.American_staffordshire_terrier',
 '009.American_water_spaniel',
 '010.Anatolian_shepherd_dog']
In [43]:
class_names[:10]
Out[43]:
['Affenpinscher',
 'Afghan hound',
 'Airedale terrier',
 'Akita',
 'Alaskan malamute',
 'American eskimo dog',
 'American foxhound',
 'American staffordshire terrier',
 'American water spaniel',
 'Anatolian shepherd dog']
In [44]:
from PIL import Image
import torchvision.transforms as transforms

def load_input_image(img_path):    
    image = Image.open(img_path).convert('RGB')
    prediction_transform = transforms.Compose([transforms.Resize(size=(224, 224)),
                                     transforms.ToTensor(), 
                                     standard_normalization])

    # discard the transparent, alpha channel (that's the :3) and add the batch dimension
    image = prediction_transform(image)[:3,:,:].unsqueeze(0)
    return image
In [45]:
def predict_breed_transfer(model, class_names, img_path):
    # load the image and return the predicted breed
    img = load_input_image(img_path)
    model = model.cpu()
    model.eval()
    idx = torch.argmax(model(img))
    return class_names[idx]
In [46]:
for img_file in os.listdir('./images'):
    img_path = os.path.join('./images', img_file)
    predition = predict_breed_transfer(model_transfer, class_names, img_path)
    print("image_file_name: {0}, \t predition breed: {1}".format(img_path, predition))
image_file_name: ./images/sample_human_output.png, 	 predition breed: Dogue de bordeaux
image_file_name: ./images/sample_dog_output.png, 	 predition breed: Greyhound
image_file_name: ./images/Labrador_retriever_06449.jpg, 	 predition breed: Flat-coated retriever
image_file_name: ./images/American_water_spaniel_00648.jpg, 	 predition breed: Curly-coated retriever
image_file_name: ./images/Curly-coated_retriever_03896.jpg, 	 predition breed: Curly-coated retriever
image_file_name: ./images/Brittany_02625.jpg, 	 predition breed: Brittany
image_file_name: ./images/Labrador_retriever_06457.jpg, 	 predition breed: Golden retriever
image_file_name: ./images/Labrador_retriever_06455.jpg, 	 predition breed: Chesapeake bay retriever
image_file_name: ./images/sample_cnn.png, 	 predition breed: Chihuahua
image_file_name: ./images/Welsh_springer_spaniel_08203.jpg, 	 predition breed: Welsh springer spaniel

Step 5: Write your Algorithm

Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,

  • if a dog is detected in the image, return the predicted breed.
  • if a human is detected in the image, return the resembling dog breed.
  • if neither is detected in the image, provide output that indicates an error.

You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and dog_detector functions developed above. You are required to use your CNN from Step 4 to predict dog breed.

Some sample output for our algorithm is provided below, but feel free to design your own user experience!

Sample Human Output

(IMPLEMENTATION) Write your Algorithm

In [47]:
### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.

def run_app(img_path):
    ## handle cases for a human face, dog, and neither
    img = Image.open(img_path)
    plt.imshow(img)
    plt.show()
    if dog_detector(img_path) is True:
        prediction = predict_breed_transfer(model_transfer, class_names, img_path)
        print("Dogs Detected!\nIt looks like a {0}".format(prediction))  
    elif face_detector(img_path) > 0:
        prediction = predict_breed_transfer(model_transfer, class_names, img_path)
        print("Hello, human!\nIf you were a dog..You may look like a {0}".format(prediction))
    else:
        print("Error! Can't detect anything..")
    
In [48]:
for img_file in os.listdir('./images'):
    img_path = os.path.join('./images', img_file)
    run_app(img_path)
Hello, human!
If you were a dog..You may look like a Dogue de bordeaux
Dogs Detected!
It looks like a Greyhound
Dogs Detected!
It looks like a Flat-coated retriever
Dogs Detected!
It looks like a Curly-coated retriever
Dogs Detected!
It looks like a Curly-coated retriever
Dogs Detected!
It looks like a Brittany
Dogs Detected!
It looks like a Golden retriever
Dogs Detected!
It looks like a Chesapeake bay retriever
Error! Can't detect anything..
Dogs Detected!
It looks like a Welsh springer spaniel

Step 6: Test Your Algorithm

In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?

(IMPLEMENTATION) Test Your Algorithm on Sample Images!

Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.

Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.

Answer: (Three possible points for improvement)

In [50]:
## TODO: Execute your algorithm from Step 6 on
## at least 6 images on your computer.
## Feel free to use as many code cells as needed.

## suggested code, below
for file in np.hstack((human_files[:10], dog_files[:10])):
    run_app(file)
Hello, human!
If you were a dog..You may look like a Basenji
Hello, human!
If you were a dog..You may look like a Dogue de bordeaux
Hello, human!
If you were a dog..You may look like a Norfolk terrier
Hello, human!
If you were a dog..You may look like a Miniature schnauzer
Hello, human!
If you were a dog..You may look like a Bichon frise
Hello, human!
If you were a dog..You may look like a Papillon
Hello, human!
If you were a dog..You may look like a German shorthaired pointer
Hello, human!
If you were a dog..You may look like a Bull terrier
Hello, human!
If you were a dog..You may look like a Dachshund
Hello, human!
If you were a dog..You may look like a Bull terrier
Dogs Detected!
It looks like a Dalmatian
Dogs Detected!
It looks like a English springer spaniel
Dogs Detected!
It looks like a Beagle
Dogs Detected!
It looks like a Boston terrier
Dogs Detected!
It looks like a Boston terrier
Dogs Detected!
It looks like a Boston terrier
Dogs Detected!
It looks like a French bulldog
Dogs Detected!
It looks like a French bulldog
Dogs Detected!
It looks like a French bulldog
Dogs Detected!
It looks like a French bulldog
In [53]:
my_human_files = ['./my_images/human1.jpg', './my_images/human2.jpg', './my_images/human3.jpg' ]
my_dog_files = ['./my_images/dog1.jpg', './my_images/dog2.jpg', './my_images/dog3.jpg']
In [54]:
## tested my pictures on the computer
for file in np.hstack((my_human_files, my_dog_files)):
    run_app(file)
Hello, human!
If you were a dog..You may look like a Bull terrier
Hello, human!
If you were a dog..You may look like a Bullmastiff
Hello, human!
If you were a dog..You may look like a Poodle
Dogs Detected!
It looks like a American eskimo dog
Dogs Detected!
It looks like a German shepherd dog
Dogs Detected!
It looks like a Golden retriever
In [ ]: